UC Berkeley

γ-rays are an indispensable probe of neutron star and black hole systems, some of the most extreme environments in our universe. The penetrating power of γ-rays allows us to probe deep within these often obscured systems. In particular, polarization measurements of neutron stars and black holes can give essential clues about γ-ray emission mechanisms, source geometry, and magnetic field structure where spectral, temporal, and imaging analysis fall short.

The Compton Spectrometer and Imager (COSI) is one of the few γ-ray telescopes designed as a polarimeter. COSI is a wide-field balloon-borne soft γ-ray (0.2-5 MeV) telescope, and one of its main goals is to measure the polarization of γ-rays emitted by compact objects. As a Compton telescope, COSI is inherently sensitive to polarization: polarized photons preferentially Compton scatter orthogonally to their polarization direction. In May of 2016, COSI was launched from Wanaka, New Zealand, on NASA’s new super pressure balloon and flew for 46 days before the flight was terminated in Peru. The Crab nebula, Cygnus X-1, and Centaurus A are among the compact objects detected during the 2016 flight.

A key step to performing imaging, spectral, and polarization analysis of the sources detected during the 2016 flight is to accurately simulate the detector response. To do so, I developed a detailed detector effects engine which applies the intrinsic detector performance to Monte Carlo simulations. With accurate simulations of the instrument in place, I developed a spectral analysis pipeline for sources detected by COSI. As needed for spectral analysis and polarimetry, I developed a background subtraction technique for broadband, persistent sources that utilizes the COMPTEL data space. I verified these new analysis methods using the COSI observation of GRB 160530A: after subtracting the background using the COMPTEL data space method and fitting the spectrum, the spectral parameters are consistent with those measured by another instrument.

I used the COMPTEL data space background subtraction algorithm to fit the Crab spectrum, but concluded that the algorithm is limited in estimating the background accurately enough in cases where the source is background-dominated. I used simulations to assess the prospects for polarimetry of COSI’s observation of the Crab and inferred that we require a better mechanism of albedo radiation background rejection to perform polarimetry of this particular observation. I note that similar observations from the COSI instrument on a satellite platform would lead to more success in measuring the spectra and polarization properties of compact objects, and that the analysis methods developed in this work could be easily applied.